You are a researcher interested in estimating the effect of…

You are a researcher interested in estimating the effect of a policy on some dependent variable Y. You are not able to conduct an actual randomized experiment to test this, but you have data on two different states, observing the Y value for each state at each of 5 time periods. In the first 4 time periods neither state has implemented the policy, but starting in the 5th time period, one state (the Treated one) implemented the policy while the other state (the Untreated one) still has not implemented the policy. You hope to use the difference in differences research design to estimate the effect of the policy on Y. The figure below shows these data: Answer the following questions based on this data.    a) What is the key assumption of the difference in differences design and does it seem reasonable here? (i.e. does the assumption seem true, at least approximately?…how can you tell from the figure?) b) What would you estimate is the effect of the policy on Y? Explain how you figured this out, referring specifically to the figure above. You only need to be approximate here – the important thing is you explain how we can learn about the policy’s effect on Y by looking at these data.